@HardwareGeek the first rule of forum etiquette is "You don't talk about forum etiquette"
Posts made by japonicus
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@antiquarian I don't think the video's a good fit for this thread because it seemed like party-political trolling, but given that it was there I thought I might as well respond to it, rather than to a meta-argument about forum etiquette.
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@antiquarian said in Tales from Coronavee-rooss Italy, mamma mia!:
@Kamil-Podlesak said in Tales from Coronavee-rooss Italy, mamma mia!:
It might not be strictly garage stuff, but it definitely does not bring anything to this, already quite big, topic.
Given that the "political theater" in question is an action that the two elected officials in the video have required of the residents of their state, and that the rationale behind the orders was the suppression of the virus that is the subject of this thread, I would strongly disagree with your assessment.
That it's 'theatre' doesn't mean it's not sincere, or that it's unreasonable. There's obviously no technical necessity to wear a mask while stood on a podium way back from a crowd, but the crowd ought to be masked and therefore the politician needs to lead by example.
Similarly, Charlton Heston's farcical rifle waving 'cold dead hands' speech was obviously theatrical as he presumably didn't anticipate a gun fight yet his belief in guns was probably quite genuine.
-
RE: WTF is happening with Windows 10? And nothing else
@Gąska My initial answer was obviously somewhat facetious and thinking about it more, there might be a narrow use case for machine learning if it allowed deployment to six-year old Moldovian PC's to be constrained until the bug could be diagnosed properly.
What irritates me is the idea that 'Machine Learning' could be anything more than a band-aid in this context. It's hardly something to boast about as it's not going to help with fixing the underlying bugs or improving the quality of future patches.
-
RE: WTF is happening with Windows 10? And nothing else
@Gąska Assuming (this being Microsoft) that nothing has been tested ahead of time: you send the update to a random sample of users and receive a slew of error reports, with some diagnostic information.
A 'machine learning' result would be: the update fails with 78% probability if the PC is more than 6 years old and that users in Moldovia have 10% less chance of success. That's not a particularly good place to start debugging from...
Instead, non-probabilistic analysis reveals that the update crashes with ATI Rage II graphics cards and/or when there is less than 1.2GB of hard disk space left.
-
RE: WTF is happening with Windows 10? And nothing else
Current status as of July 22, 2020
Windows 10, version 2004 is available for users with devices running Windows 10, versions 1903 and 1909, who manually seek to “Check for updates” via Windows Update. We are now starting a new phase in our rollout. Using the machine learning-based (ML-based) training we have done so far, we are increasing the number of devices selected to update automatically to Windows 10, version 2004 that are approaching end of service. We will continue to train our machine learning through all phases to intelligently rollout new versions of Windows 10 and deliver a smooth update experience. The recommended servicing status is Semi-Annual Channel.
@Gąska said in WTF is happening with Windows 10? And nothing else:
@topspin that part actually makes sense. They're making statistics on what kinds of problem what kinds of PC configurations have, calculate compatibility score based on this, make tiny patches to fix things, then deploy W10 to new computers that are predicted based on their spec to have high compatibility score. Machine learning is a perfect tool for that.
Machine learning seems a lazy and appalling fit for this task. An update will either fail or succeed and for a well-defined and explicable reason - there are no probabilistic vagaries involved and, if there's any hope of patching them, reasons-for-failure must be precisely understood, and therefore exactly testable for. The criteria for blocking an update are therefore going to be similarly well-defined.
In this context the whole 'machine learning' thing is marketing bullshit.
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@boomzilla said in Tales from Coronavee-rooss Italy, mamma mia!:
@Grunnen sometimes. But the way the world works is that we have to get to herd immunity, so the alternative to getting relatively safe people infected is to get relatively vulnerable people infected, which is what you seem to be arguing for, but I think that in reality you're just not thinking it through enough, TBH.
Obviously, when you have multiple generations living in the same household, that's a big problem, but it's not a universal situation and treating the entirety of society that way is really bad.
I don't think it's possible to adequately identify who is at risk or even where risk can be defined to isolate only those individuals - once you try, the bubble of linked people very rapidly expands to a scale close to complete lock-down.
A few specific examples:
I would normally share an office with a 40-something guy who, both in terms of an underlying health issue and ethnic origin is at greater risk. He has two teenage children and his wife works in a public facing role. There's no way he can shelter if society re-opens fully.
A friend in his 50's who is notably fit and healthy (i.e. nothing to suggest being at risk) caught the virus in March while on a skiing holiday. He ended up in hospital on oxygen for several days, (his partner was affected more severely). Both fully recovered, but if that had happened at a point when hospitals were overwhelmed then the outcome would probably have been different.
One of my colleagues (healthy mid-30's woman) was off-work for a month with the virus - if that happened widely then sick-leave would rapidly bring society to a halt.
It's not common among urban populations in the UK for three generations to be living in the same cramped house (notably, but certainly not exclusively, Indian sub-continent communities). That was the case where I lived in Manchester - where the grand-parent generation were typically strikingly unhealthy (obese, probably malnourished and often diabetic); the middle generation worked at the local hospital (a major employer in the area) and the children went to school.
My point in this long-winded post is that you either can't identify who is at risk, or even if you can then often (usually) you can do little to protect them. The broad-brush approach of limiting viral spread across the whole of society is the only thing that's viable.
'Herd immunity' can be achieved with widespread vaccination - in which case the aims of social controls only need to be to hold-off widespread infection for long enough to allow a vaccine to be developed.
-
RE: Having problems with unit testing philosophy.
@Gąska said in Having problems with unit testing philosophy.:
@xaade said in Having problems with unit testing philosophy.:
So,
bob.JumpAndRun();
test if bob called jump
test if bob called runactual code
bob.JumpAndRun() { Jump(); Run(); }
Like I said, complete tautologies, why is this being done?
Accidents happen. One day someone might modify the code to do just a bit more logging and accidentally disable the call to
Run
.More likely, someone will refactor the code and misunderstand what JumpAndRun was meant to achieve - this test won't catch that.
Rather than caring about Running and Jumping it might be better to have separate tests for the desired side effects:
Bob moves from A to B
Bob becomes more fit
Jumping must happen before running
Rita and Sue are satisfiedThis avoids problems caused by the well-meaning programmer who puts Bob on a treadmill.
-
RE: Update on Administration & Community Changes
@MrL said in Update on Administration & Community Changes:
@HardwareGeek said in Update on Administration & Community Changes:
A couple of times a certain person known for PWI
Don't throw us all into one basket.
I thought PWI was the norm around here
-
RE: Reserved
I usually carry a backpack, but it's not socially acceptable in other settings than 100% casual
E_NO_BACKPACK does not compute.
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
Status: Am I a horrible person if I don't wear a mask?
{for context: live in UK, where crap politicians have been sending out completely ambiguous messages corona-wise.}
So far as I can tell, everyone should be wearing a mask in public places, to reduce spread of the plague to innocent bystanders (but not for personal protection).
Simple. The problem is that when I visit a local supermarket then, right now, no-one (neither staff nor customers) is masked and, on past experience, the odd (but possibly ethical) mask wearers are shunned as though they are typhoid-Mary's.
Several times now I've walked to the supermarket with a mask in my pocket, quite prepared to wear it, but baulked at the last moment because no-one else is wearing one.
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@Mason_Wheeler said in Tales from Coronavee-rooss Italy, mamma mia!:
Look up "pleurisy" for one of the more common ugly side-effects of excessive mask wearing.
Citation needed. (and not the story currently doing the rounds on social media)
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@lolwhat said in Tales from Coronavee-rooss Italy, mamma mia!:
This motherfucking virus has been around a very long time, and the Chinese did everything they could to hide it.
How long does it usually take to put the jigsaw pieces together and spot the emergence of a new virus? The time-span involved doesn't suggest to me a cover-up. It's easy with hindsight to complain about failings, probably much harder in reality to spot a new virus.
Hospitals would have seen an up-tick in elderly and infirm patients in a heavily polluted city presenting with respiratory problems at a time of year when flu would also be beginning to kick in. Most of the searches for coughs and diarrhoea would be from people who never sought medical attention (or, in China, went to a traditional medicine quack who probably wouldn't log or pass on samples for testing), so most of that could easily fly under the radar.
Would you also argue that Saudi Arabia covered-up MERS? The first case of that is thought to have been June 2012, but it wasn't identified as a novel virus and flagged to the world until mid-September.
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@izzion said in Tales from Coronavee-rooss Italy, mamma mia!:
@japonicus
Deplorable was crowned the official word for that in 2016.TIL, good to know the cromulent lexicon.
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@boomzilla said in Tales from Coronavee-rooss Italy, mamma mia!:
@Rhywden said in Tales from Coronavee-rooss Italy, mamma mia!:
Do you really think that this
despicabledeplorable "Redstate" site is worthy of your attention?Words matter.
Yes they do, but in this case 'despicable' seems fine
-
RE: The Official Status Thread
@Gąska said in The Official Status Thread:
Why would you want your email client to organize mail in a way you can't comprehend?
Plausible deniability
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@loopback0 said in Tales from Coronavee-rooss Italy, mamma mia!:
UK lockdown has been relaxed slightly.
-
RE: Tales from Coronavee-rooss Italy, mamma mia!
@Benjamin-Hall said in Tales from Coronavee-rooss Italy, mamma mia!:
@topspin no. The main models include strong restrictions. And have since the beginning.
The 465 000 figure, which appears to be from a Stanford University model, was the prediction with no social distancing, at the same time they were predicting 18 000 if shelter-in-place was applied (still way out, but there were many unknowns).
Can't find original study but there's a news report from March here:
With limited action, Florida's hospitals would reach an overload of over 465,000 patients by April 24, the model shows. With three months of social distancing, Florida would still see 185,000 hospitalizations by May 14. With mandatory sheltering in place, the need is reduced to 18,000 hospitalizations by July.
-
RE: Reframe interview questions to assess skills, not stories
The second set of questions is easier to answer, but possibly less informative about the candidate's capabilities.
If interviewing someone for a senior-ish role then it doesn't seem unreasonable to ask for examples of things they've actually done rather than hand-waving hypothetical situations. Technical skills are likely to be tested later on anyway.If the candidate can't think of an exact example (either because they haven't ever done precisely that thing or just because interviews are horrible and fry brains) it should still be possible to come up with some sort of partial example or explanation. Talking about something real is going to give away much more about how the candidate actually works and behaves (e.g. does it seem like they're taking credit for someone else's work?)
Maybe a competent interviewer (don't think I've ever met one) ought to use a mixture of the two approaches.
-
RE: The Official GDPR Lawsuit thread
@JBert said in The Official GDPR Lawsuit thread:
Digging through https://transparencyreport.google.com/ might have up-to-date numbers, but for now.
Worth reading. Was surprised that Google does actually deny half (54%) of RTBF requests.
Contrary to some of the assertions in this thread, only 6.4% of requests relate to crime (but add to that 6.3% for 'professional wrongdoing').
Google lists a large number of examples of how how requests are dealt with - most look sensible. Some of the criminality that's covered is a bit disturbing, but is handled in a way that appears consistent with spent-conviction rules (which makes sense).
-
RE: The Official GDPR Lawsuit thread
@Mason_Wheeler said in The Official GDPR Lawsuit thread:
If people understand that things they do wrong are not going away, you'll likely end up seeing less petty theft.
I completely disagree with that sentiment, but in any case it misses the point that the GDPR provisions are not just (or evenly mainly) about literal criminality, but a wider legal effort to take steps against deep profiling of the minutiae of your life (all the non-criminal aspects).
Do you want every future employer, romantic partner, banker, policeman etc to know how long you spend on TDWTF (or on less salubrious parts of the internet); that you were a brat in high school; your views on religion; who you retweeted; what fancy dress costume you wore when you were 15; everyone you ever dated; every message board post (classified by subject, emotional intent and literacy level)?
If you found out that someone was compiling such a profile (and maybe selling it) you'd probably want to stop them - to not be 'remembered' in that level of detail.
Many countries have laws against compiling particular forms of employee blacklists (records of union activity etc) but, given modern technical capabilities, the problem now is that a profile of your entire life history could be compiled trivially and completely beyond your control. There is a legal need to be able to constrain that. Some of the GDPR provisions could be seen as an imperfect first attempt.
Masking a few embarrassing google searches is a side-effect (and as has already been noted, predates GDPR anyway).
-
RE: The Official GDPR Lawsuit thread
@Mason_Wheeler said in The Official GDPR Lawsuit thread:
@dfdub said in The Official GDPR Lawsuit thread:
You replied to each of the three sentences in this paragraph individually, which doesn't make any sense. The statement I'm making is that Google's monopoly is a large part of the problem.
Which I already dealt with: there is no Google monopoly! A false premise implies anything. Given that the rest of the paragraph rests on this premise, it makes all the sense in the world to deal with it individually!
When google has appears to have 93% share of the UK search market that seems like a defacto monopoly.
That's relevant because it means there is no commercial incentive for Google to bother fighting spurious take down requests. If a credible competitor did fight tooth-and-nail against every RTBF request then reputational damage to Google would force it's hand. As it is, nothing gets legally test so there is little clear definition of what counts as a legitimate right-to-be-forgotten (and there are many legitimate reasons).
-
RE: The Official Status Thread
@Zerosquare turning off-and-on again is much more reliable.
-
RE: The Official Status Thread
Status just photographed someone's ID card (on a wooden table) as legally required proof of right-of-residence. Really hate living in a proto-fascist state co/ Boris Johnson etc.
-
RE: Just learn to code
@sh_code said in Just learn to code:
Those who stayed say they were given vague assignments with little instruction and told to "Google it" when they had questions"
"Google it" isn't teaching. Of course a lot of programming ends up like that, but it's not good to turn out more stackoverflow copy-pasta.
Vague assignments are crap in this context. If active teaching is involved then the process of solving the problem ought to start to with 'how' not 'what' - otherwise you're wasting the time of the student.
I'd call that a solid first lesson in programming.
No. It's what a lot of real-life is like, but it has no place when it comes to teaching programming.
And I was 10 years old
The way you learn when you're ten is radically different from the approach when you are older. Like you, I started learning programming when I was young (~ 8/9) - at that point I'd absorb details through my skin without conscious effort - it's not like that now
-
RE: THE BAD IDEAS THREAD
@Rhywden said in THE BAD IDEAS THREAD:
@jinpa said in THE BAD IDEAS THREAD:
@Rhywden said in THE BAD IDEAS THREAD:
If the people in the program knowingly file a false violation, they will be given a civil fine of up to $100.
I can see why you highlighted that. Aren't there criminal (as opposed to just civil) laws already in place against that sort of thing?
Since parking in the wrong spot is not a crime, I don't see why that should invoke criminal laws instead of civil liabilities.
Malicious false accusations sound criminal not civil (in the same vein as 'false arrest' etc.)
The bigger issue is that, being USA, the neighbours will just start shooting each other.Apart from that the rest of the proposed rules re. pedestrian/bike lane provision sound very sensible.
-
RE: WTF Bites
Amazon now does time travel:
(bear in mind that the status message switched to this some time after 8pm on Tuesday when the package was still hundreds of miles away at the wrong depot)
-
RE: Patented Pepsi Potato Prosecution
@Gąska said in Patented Pepsi Potato Prosecution:
Okay, let's say they only traded "farm produce including seed of a variety", and not the original branded seed. How the hell were the neighbors able to plant many acres of a sterile plant?
'Seed' as an agricultural reference to potatoes means tubers - vegetative rather than sexual reproduction - so the sterility of the plants is immaterial.
-
RE: Patented Pepsi Potato Prosecution
@Benjamin-Hall you're taking the norms of a highly industrialised agricultural production system in a wealthy country and trying to apply the same to a very different situation, where normal practice might often be to replant seed (or in this case tubers) from a previous crop.
I don't dispute that the idealised approach using well screened commercial stock will produce a better outcome but, morally, trying to impose the legal framework of a rich industrial county on the developing world is flawed.
For those who can, there are already probably economic incentives to purchasing disease free seed each year, so this situation can already be resolved without intervention. You can have a situation where farms can choose top quality certified seed and sell their crop at higher prices or save their own seed and probably take a hit with a poor grade crop.
(though looking at the crap state of potatoes sold for food in the UK even fully industrialised production is leading to a lot of disease at the moment)
-
RE: Patented Pepsi Potato Prosecution
@Benjamin-Hall Well your experience is different from mine then.
-
RE: Patented Pepsi Potato Prosecution
@Benjamin-Hall said in Patented Pepsi Potato Prosecution:
You are reproducing a patented product, therefore you are in violation of the patent. Full stop. Buying a copy does not give you the right to create more of it.
I'd be a bit miffed if I bought a seed potato only to discover that I wasn't allowed to let it grow.
Especially for potatoes that do not reproduce by seed, like the patented ones. There you have to cut them up, plant the eyes, and go from there. It cannot happen naturally.
You have evidently never grown potatoes. After a crop has been harvested, the following year there will always be new potato plants growing from the ones that got missed. No need for human intervention whatsoever.
-
RE: Reality or WTF?
@dkf said in Reality or WTF?:
That depends on what the database engine does with caching.
True, but neither MySql or Postgres cache prepared statements beyond the end of a session and those two databases probably account for the vast majority of php based setups.
-
RE: We need to be more user hostile to help them embrace freedom!
@Tsaukpaetra said in We need to be more user hostile to help them embrace freedom!:
Didn't you know? Murder investigations merely increase tolerance for murder. 🔪
Probably true to some extent. All sorts of nuanced classifications of homicide have been invented along with several different defences, whereas historically murder was murder and hang-em-all (unless you were clergy, sufficiently wealthy or royal...)
(off-topic: it seems quite apt that the suggestion for :murder: is a red-coat solider... )
-
RE: ⛑ Relative Oriented Programming
@gleemonk rewriting it now would be a kick in the teeth for your relative. Effectively you'd be taking over his project - because the learning curve for him to pick it up again would be very steep and demoralising.
I've been in a similar position to you and quickly learnt to bite my lips and make minor tweaks rather than undermine someone who'd learnt a huge amount from scratch but, through inexperience, had produced a monstrosity (mostly just for her own personal use).
Offer specific guidance when asked but otherwise stay well clear.
-
RE: Reality or WTF?
@Captain said in Reality or WTF?:
@Tsaukpaetra OK, then tell them prepared statements are 3-17% faster, for free. The database gets to prepare, and it saves time. That sounds like something people in a PHP shop would believe. (Just assuming PHP based on how amateurish they sound)
For PHP I suspect that much of the obsession with prepared statements is misplaced cargo-cultism: a reflection of people confusing parameterized queries (good) with prepared statements (mostly irrelevant or counter-productive in the context of a short-lived php process).
There's no point creating a prepared statement if the process uses it only once and then throws it away, which would be typical usage from PHP.
-
RE: Reality or WTF?
@dkf said in Reality or WTF?:
I'd expect it on virtually all of them. Not needing to reparse the SQL at all (because it's character for character identical to a previous run) is almost always going to be a bit cheaper option than guessing that
$THIS
is sufficiently similar to$THAT
…Does sql parsing (as distinct from query planning) take a significant amount of time, compared to the time taken to plan and then run the query?
-
RE: Using Java? Google thinks you might be suicidal
@PleegWat said in Using Java? Google thinks you might be suicidal:
Trying some variants, this seems to be tied to the
how to tell if you're running X
prefix. Filling in a number of different similar words for thejdk or jre
part gives the same suicide link at the top (though wildly different results elsewhere).For me 'jre' appears is significant:
how to tell if you're running jre
triggers
but nothow to tell if you're running jdk
even weirder, the contraction matters:
how to tell if you are running jre
doesn't workWhat other meanings are there for JRE that work across multiple languages?
-
RE: In other news today...
@izzion said in In other news today...:
@topspin
Except (1) the lead opposition party is in favor of Remain and (2) most of the 105 MPs are on the extreme Leave side of the spectrum. They voted against the deal because it wasn’t a hard ENOUGH Brexit, and would prefer a No Deal exit over Remain.But the leader of the opposition secretly wants to leave (because he's dreaming of a communist utopia) despite the rest of his party remaining, while the prime-minister originally pretended to want to remain but actually wants to leave (mostly because she's still sulking about some European Court of Justice rulings that went against her).
So at least half the country wants to remain (but many were too lazy to vote), ~ 2/3rds of parliament want to remain, but a couple of idiot party leaders are spoiling everything and will throw the toys out the pram rather than talk to each other.
-
RE: WTF Bites
@Lorne-Kates said in WTF Bites:
No, all he needs is a treatment with the right essential oils. They will cure him.
Well duh, people are complex, like cars. Unless you change the oil regularly things break and top quality essential is obviously better and more oily than regular fat.
-
RE: In other news today...
@PJH said in In other news today...:
Seems UK plod should be allowed nothing more technical than a whistle and a truncheon.
Scary club-wielding suspect seen with un-monitored communication device === terrorist.
And possibly a (plastic) knife.
+ knife ( )
-
RE: The "but what does it do" effect: how often do you have trouble understanding what stuff does?
@anonymous234 said in The "but what does it do" effect: how often do you have trouble understanding what stuff does?:
I've been thinking about starting a wiki dedicated to describing software products in a simple way - from the point of view of what "services" it provides for the user or other software.
Too often there's a reason for the sparse description: the emperor has no clothes.
-
RE: The Official Status Thread
Status: social media (was forced to join whatsapp and reply to ppl on twitter)
-
RE: The Whisky Topic
@dkf said in The Whisky Topic:
@Gribnit said in The Whisky Topic:
England is tending towards
sherryginFTFY
-
RE: WTF Bites
@kazitor what's the word for legible but for audio?
'intelligible' could be used, though not specific to audio and plenty of podcasts aren't to begin with.
-
RE: In other news today...
@Karla said in In other news today...:
@boomzilla said in In other news today...:
why wouldn't they do an ultrasound to date the pregnancy the first time they told her she was pregnant?
From the article it sounds as though she may have kept taking pregnancy tests at home and only got a scan after she finally sought medical advice:
After blacking out in her office admin job, Ms Favell plucked up the courage to see a GP in 2016, who said she must be pregnant despite blood negative tests.
...
Still believing Ms Favell to be pregnant, her GP referred her for an ultrasound scan in January last year.The GP probably wouldn't have had a scanner and getting a hospital appointment for a scan would take a few days, (or maybe slightly longer over Christmas???) - not an issue in a normal case. That assumes that Favell co-operated straight away and wasn't still just in denial.
The crazy thing is that she went for so long without recognising that something was seriously wrong (or more likely she knew but was scared of confirmation).
-
RE: WTF Bites
@sweaty_gammon It's always easier to make a complex architecture. Simplicity is hard, and requires a thorough understanding of both the domain and tools used to be done correctly.
Complexity also allows people to spend a lot of time and lines-of-code on architectural busy-work, building abstractions for their abstractions and putting off tackling (or burying) all the complex bits that require some understanding.
-
RE: The latest npm security kerfuffle
@izzion said in The latest npm security kerfuffle:
@japonicus
But isn't the time and effort spent part of the efficiency as well?Yes, I'm definitely not advocating that everything should be written in low-level languages - performance often doesn't matter. I was responding to @Gąska who I understood to be arguing narrowly in terms of code-performance rather than efficiency of a project.
@Gąska said in The latest npm security kerfuffle:
It's a null hypothesis. The burden of proof is on the other side.
I'll conceded the point on pedantry, but still don't remotely accepted the lack of correlation which goes against common experience.
@Gąska said in The latest npm security kerfuffle:
And what about average programmers? You know, the kind that writes 90% of the code in the world?
I suspect (though you might dispute it ) that use of low-level language also has some correlation with expertise. Inexperienced (or even average) programmers are far less likely to work in low-level languages professionally. That most code is written in high-level languages (or that many programmers are less than competent) doesn't affect the correlation between language-level and raw code efficiency.
-
RE: The latest npm security kerfuffle
@Gąska said in The latest npm security kerfuffle:
There is no correlation between abstraction level and performance.
That's a very bold and improbable general claim - do you have any evidence? Optimisation techniques are getting better, but I'd be very surprised if a skilled programmer couldn't usually achieve greater efficiency in a low-level language (if they spent the time and effort).